Physical therapies for reducing and controlling lymphoedema
of the limbs (Review)
Preston NJ, Seers K, Mortimer PS
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library
2008, Issue 3
http://www.thecochranelibrary.com
Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
T A B L E O F C O N T E N T S
1HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
21DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
22AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
22ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
22REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
24CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
33DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
33APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
34WHAT’S NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
35HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
35CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
35DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
35SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
36NOTES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
36INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
iPhysical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
[Intervention Review]
Physical therapies for reducing and controlling lymphoedemaof the limbs
Nancy J Preston1, Kate Seers1, Peter S Mortimer2
1Research Team, RCN Institute, Oxford, UK. 2Division of Physiological Medicine, Cardiac and Vascular Sciences, St Georges Hospital
Medical School, London, UK
Contact address: Kate Seers, Research Team, RCN Institute, Radcliffe Infirmary, Woodstock Rd, Oxford, OXON, OX2 6HE, UK.
Editorial group: Cochrane Breast Cancer Group.
Publication status and date: New search for studies and content updated (no change to conclusions), published in Issue 3, 2008.
Review content assessed as up-to-date: 19 February 2008.
Citation: Preston NJ, Seers K, Mortimer PS. Physical therapies for reducing and controlling lymphoedema of the limbs. Cochrane
Database of Systematic Reviews 2004, Issue 4. Art. No.: CD003141. DOI: 10.1002/14651858.CD003141.pub2.
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
A B S T R A C T
Background
Lymphoedema is the accumulation of excess fluid in the body caused by obstruction of the lymphatic drainage mechanisms. Management
involves decongesting the reduced lymphatic pathways in order to reduce the size of the limb. There is a great deal of debate as to which
components of a physical treatment programme are the most crucial.
Objectives
To assess the effect of physical treatment programmes on:
volume, shape, condition and long-term control of oedema in lymphoedematous limbs;
psycho-social benefits.
Search strategy
We searched the Cochrane Breast Cancer Group trials register (October 2007), the Cochrane Central Register of Controlled Trials
(The Cochrane Library Issue 1, 2008), MEDLINE, EMBASE, CINAHL and the National Research Register (February 2008) and
UnCover, PASCAL, SIGLE, reference lists produced by The British Lymphology Society and The International Society of Lymphology
congress proceedings (September 2003).
Selection criteria
Randomised controlled clinical trials that tested physical therapies with a follow-up period of at least six months.
Data collection and analysis
Two blinded reviewers independently assessed trial quality and extracted data . Meta-analysis was not performed due to the poor quality
of the trials.
1Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Main results
Only three studies involving 150 randomised patients were included. Since none studied the same intervention it was not possible to
combine the data. One crossover study of manual lymph drainage (MLD) followed by self-administered massage versus no treatment,
concluded that improvements seen in both groups were attributable to the use of compression sleeves and that MLD provided no extra
benefit at any point during the trial. Another trial looked at hosiery versus no treatment and had a very high dropout rate, with only 3
out of 14 participants in the intervention group finishing the trial and only 1 out of 11 in the control group. The authors concluded
that wearing a compression sleeve is beneficial. The bandage plus hosiery versus hosiery alone trial, concluded that in this mixed group
of participants bandage plus hosiery resulted in a greater reduction in excess limb volume than hosiery alone and this difference in
reduction was maintained long-term.
Authors’ conclusions
All three trials have their limitations and have yet to be replicated, so their results must be viewed with caution. There is a clear need
for well-designed, randomised trials of the whole range of physical therapies if the best approach to managing lymphoedema is to be
determined.
P L A I N L A N G U A G E S U M M A R Y
Physical therapies for reducing and controlling lymphoedema of the limbs
Lymphoedema is the build up of excess fluid in the body tissues because of obstruction of lymphatic drainage back into the bloodstream.
The affected limb becomes swollen, distorted in shape with pain, discomfort all of which impair movement and daily activities. It
can be caused by a congenital abnormality, chronic venous insufficiency, damage to the lymphatic system following treatment of
cancer or filariasis, a parasitic infection endemic in parts of India and Africa. Skin care is important as the affected tissues gradually
thicken and are susceptible to inflammation and infections. People are also encouraged to exercise regularly and control their weight.
Different physical treatments aimed at improved lymph drainage include multi-layer bandaging, manual lymph drainage (MLD), self-
administered massage and compression sleeves or hosiery.
The authors of this review, which aimed to assess the effect of physical treatment programmes on the long-term control of lymphoedema,
identified only three controlled trials for inclusion. These randomised a total of 150 adults to different levels of physical treatment. One
trial involved 42 women with unilateral lymphoedema of the upper limb following treatment for breast cancer. One group received
eight sessions of MLD in two weeks and training in self-massage and both this group and the control group wore flat-knit compression
sleeves. The reductions in excess arm volume and symptoms were similar in the two groups.
A second trial involved 25 women from a local follow-up breast clinic. They were trained in self-administered massage and randomised
to wear an elastic compression sleeve or no additional treatment. The dropout rate was high, particularly in the control group, although
the authors concluded that wearing a compression sleeve was beneficial. The third trial involved 83 mostly female participants from a
lymphoedema clinic. Around two thirds had upper limb oedema. They were all taught self-administered massage. One group received
a 19-day bandaging course before being fitted with hosiery. The other group wore hosiery from the start of the trial. The reduction in
excess limb volume was consistently greater in those who started with multi-layer bandaging.
All three trials had methodological limitations, and as their data could not be combined, and they recruited only small numbers of
participants, questions relating to the effect of this type of treatment could not be answered by this review.
B A C K G R O U N D
Lymphoedema is a chronic and progressive condition resulting
from an abnormality of, or damage to the lymphatic system. Any
reduction in the capacity of the lymphatic system to drain fluid
from the interstitium and return it to the blood circulation will
cause fluid to build up in the skin and subcutaneous tissues of the
affected part of the body (Levick 1991; Mortimer 1995).
Lymphoedema has many causes but the main ones are:
2Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
• Cancer and its treatment - leading to secondary
lymphoedema;
• Congenital abnormalities of the lymphatic system - so
called primary lymphoedema;
• Chronic Venous Disease of the lower limb - lympho-venous
oedema;
• Filariasis, a parasitic infection - leading to secondary
lymphoedema
The incidence of lymphoedema following breast cancer treatment
is difficult to assess due to differences in assessments of diagno-
sis, measurement and follow up time. Petrek 1998 and Erickson
2001 conducted systematic reviews to assess the incidence of breast
cancer lymphoedema. The included studies assessed the incidence
in Europe, Australia and North America. Using different search
strategies Petrek 1998 found eight studies which included inci-
dence and Erickson 2001 found ten including potentially two
studies which were too recent to include in the review by Petrek
1998. Four papers were included in both reviews although one pa-
per was judged to be retrospective by Petrek 1998 and prospective
by Erickson 2001. Three methods of assessing lymphoedema were
included; volume, limb circumference and self-report. The time
scale used for follow up also varies. Petrek 1998 report incidence
figures of 6 to 30% however the table used to explain these figures
is unclear. Only papers which covered axillary dissection were used
in the review by Erickson 2001 and a range of 2.4 to 56% is found.
The range in follow up, if details were available at all, seems to re-
sult in this wide variety in findings. The figures for lymphoedema
of the lower limb are even less reliable but it appears to be a ma-
jor problem. In many other cancers (for example melanoma, soft
tissue sarcoma and pelvic tumours) the treatment often compro-
mises lymphatic drainage routes; chronic venous insufficiency is
also a major contributory factor in the development of chronic
lower limb oedema, as is filariasis, a parasitic infection endemic
in parts of India and Africa. Moffatt 2003 surveyed healthcare
providers in South West London and found a crude prevalence of
lymphoedema from any cause of 1.33 per 100,000. The incidence
of arm oedema was much higher in women, which reflects the
large number of women developing lymphoedema following the
development of breast cancer.
Lymphoedema can result in significant physical and psychological
morbidity. Swelling causes a disproportion in the size of a part of
the body and as such can interfere with mobility and affect the suf-
ferers’ perceptions of themselves (Tobin 1993). In addition to an
increase in size, the affected subcutaneous tissues gradually thicken
and fibrose forming a solid component to the swelling (Foldi
1985; Mortimer 1995). Pain and discomfort feature among the
physical problems associated with lymphoedema (Badger 1988;
Carroll 1992), as do recurrent attacks of infection/inflammation (
Mortimer 1995); the latter are a result of reduced local immunity
in the affected part of the body. The shape of the limb can become
distorted (Badger 1997). The impact of these physical and psy-
chosocial difficulties on the patients’ quality of life has attracted
little research interest. As things stand, the pre-morbid state has
not been identified so that the prevention of lymphoedema is not
yet an option.
The management of this condition involves decongesting the re-
duced lymphatic pathways in order to reduce the size of the limb;
encouraging the development of collateral drainage routes and
stimulating the function of remaining patent routes so as to con-
trol the swelling long-term (Foldi 1985; Mortimer 1995).
Traditionally, treatment in continental Europe has followed a
“two-phase” approach. In the first phase a collection of different
physical treatments are employed simultaneously for a course of
daily treatment; these usually include:
• Multi-layer bandaging
• Care of the skin
• Manual lymph drainage (MLD)
• Exercises to promote lymph drainage.
• Pneumatic compression (is sometimes also included here).
This phase aims to reduce the size of the limb, reverse any distor-
tion in the shape of the limb and any hardening of the subcuta-
neous tissues and improve the health of the skin.
In the second phase of treatment patients are usually required to:
• Wear strong compression hosiery to maintain the reduction
in swelling;
• Carry out regular daily exercise;
• Have regular MLD, where possible.
The aim here is to maintain the improvements gained in the first
phase.
There is a great deal of debate as to which components of a physical
treatment programme are the most crucial and whether bandages
are more effective at reducing oedema than compression hosiery.
Outside continental Europe the lack of experienced therapists and
inadequate resources mean that standard treatment is likely to con-
sist of compression hosiery with advice on skin care and exercise.
Where MLD is not available, patients are often taught to perform
a simplified form known as “simple lymph drainage” (SLD) or
“self-administered massage” (SAM) however as to which of these
methods is the more effective, and what effect they have is the
subject of much speculation by therapists. Pneumatic compres-
sion therapy (PCT) used both as a way of reducing oedema and
3Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
of controlling it, is an approach on which opinion is also divided.
While PCT has been demonstrated to reduce swelling, the way
in which it does so and the possibility of the rapid displacement
of fluid to elsewhere in the body have caused concern. PCT does
not obviate the need to contain the limb with hosiery and it is
not clear that it provides anything which movement, whilst wear-
ing a short-stretch bandage or compression hosiery, does not also
provide. The role of weight control and regular exercises in the
management of lymphoedema is also thought to be important.
O B J E C T I V E S
1. To assess the effect of physical treatment programmes on the
volume, shape, condition and long-term (six months) control of
oedema in lymphoedematous limbs.
2. To assess the psycho-social benefits of physical treatment.
M E T H O D S
Criteria for considering studies for this review
Types of studies
Only randomised controlled trials were included in this review. If
insufficient details were provided as to the randomisation proce-
dure, authors were contacted for more details. However, due to the
lack of detail obtained in relation to the randomisation method,
those trials described as randomised were included in the review.
Types of participants
Studies had to recruit adults (greater than 18 years of age) with a
diagnosis of lymphoedema defined as clinically detectable oedema
from a peripheral cause of greater than three months duration.
Follow up had to be for at least six months. In patients with uni-
lateral oedema, the increase in limb volume of the swollen limb
had to be at least 10% above that of the contra-lateral normal limb
volume. This cut-off point was based on information regarding
the per cent difference in volume between left and right limbs in
the normal population, which can be as high as 8 to 9% (Sitzia
1997). All types of lymphoedema were included, non cancer-re-
lated and cancer- related. Cancer patients had to have completed
their cancer treatment at least six months before entering the trial
and could not have evidence of recurrent malignant disease when
going into the trial. If multi-layer bandaging was the intervention
being studied then the participants could not have already received
a course of treatment involving bandaging in the six months pre-
ceding their entry into the trial.
Types of interventions
• Multi-layer bandaging compared to hosiery
• Hosiery/multi-layer bandaging compared to exercise/no
treatment
• Hosiery/multi-layer bandaging compared to MLD/
SLD/SAM
• MLD compared to SLD/SAM
• MLD/SLD/SAM compared to exercise/no treatment
• Exercise compared to no treatment
• PCT compared to hosiery/Multi-layer bandaging/SLD/
SAM/exercise
• PCT compared to no treatment
Types of outcome measures
Main outcomes
1. Volume measurement of limbs - reports of circumference mea-
surements alone were not sufficient unless taken at 4 cm intervals
when they could be converted to volume measurements by the for-
mula circumference squared over Pi. Volume could be measured
by:
• water displacement;
• electronic volumeter;
• calculated from surface measurements.
2. Where lymphoedema was unilateral the normal limb should act
as the patient’s own control - volume had to be expressed as the
excess limb volume over the normal limb volume; any reduction/
increase should be reported as the percentage reduction/increase
in the excess limb volume.
3. Impact on quality of life.
4. Impact on patient’s sense of well-being.
5. Impact on patient’s mobility.
6. Reduction in recurrent infections.
Any psychological benefit such as changes in a sense of well-being/
quality of life or improvement in body image had to be reported
using validated scales. Other aspects of physical morbidity such
as an improvement in the condition of the skin and the quality
of the tissues in the limb, improvement in the shape of the limb
and increased mobility had to be assessed objectively whenever
possible. Long-term and short- term adverse effects related to the
interventions were noted.
Search methods for identification of studies
The Cochrane Breast Cancer Specialised Register contains the re-
sults of searching a wide list of databases together with handsearch-
ing of specialised journals and conference proceedings. For this
updated version of the review, the authors decided not to continue
to search some databases previously searched, as the trials regis-
ter is likely to contain all relevant trials. However, supplementary
4Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
searching for this updated version of the review was undertaken
of the larger electronic databases separately by the authors as an
additional check.
Databases searched for this current updated review:
• Cochrane Breast Cancer Specialised Register
For the first full version of this review (New Reference), the Spe-
cialised Register maintained by the Cochrane Breast Cancer Group
was searched in September 2003 (details of search strategies used
by the group for the identification of studies and the procedure
used to code references are outlined in the group’s module http://
www.mrw.interscience.wiley.com/cochrane/clabout/articles/
BREASTCA/frame.html). Studies coded as ’lymphoedema’ on the
Specialised Register were extracted for consideration. This search
was repeated in October 2007 for this update.
• Cochrane Central Register of Controlled Trials (Issue 1,
2008)
See Appendix 1
• CINAHL (1982 to February 2008)
See Appendix 2
• MEDLINE (1950 to February 2008)
See Appendix 3
• EMBASE (1980 to February 2008)
See Appendix 4
• National Research Register (NRR) (February 2008)
Potentially relevant studies from this search were extracted for
consideration and attempts made to locate the authors.
See Appendix 5
Databases searched for original review only (Badger C, 2004):
• BNI (September 2003)
• CANCERLIT (September 2003)
• British Library Index (September 2003)
• UnCover (September 2003) (“UnCover Title and Subject
Lists” is a database listing journal contents pages)
• PASCAL (September 2003)
Search terms used for the above databases are included in Appendix
6. Searches went back, where possible, as far as the 1960s. In order
to identify the “grey literature” SIGLE was also searched.
Other sources searched for original review only (Badger C,
2004):
• International Society of Lymphology biennial congress
proceedings (hand searched September 2003)
• International experts in the field were contacted to see if
they held any unpublished data, as were those found to be
presenting relevant papers in conference proceedings.
All 353 members of the International Society of Lymphology were
contacted by letter to ask if they had any results of past or ongoing
studies that could be considered for inclusion in the review. Only
30 replies were received and none of these uncovered data that
could be included in this review.
Data collection and analysis
Selection
One reviewer scanned the titles and abstracts of the papers found
through searches (CB). Those clearly not relevant based on read-
ing of the abstract (for example if it was clear that they did not de-
scribe studies, or did not relate to lymphoedema) were excluded; if
no abstract was available or if it was not immediately clear that the
paper was not relevant then the full publication was retrieved for
closer review. A second reviewer (NP) looked at a sample of those
discarded to check that nothing relevant had been lost. This pro-
cess was independently undertaken by two reviewers in the 2008
update (AL and JB) and differences resolved by consensus.
Two reviewers (CB and NP or AL and JB) independently read
all the retrieved papers under consideration and independently
assessed their eligibility for inclusion according to the criteria set
out for trials above. An eligibility form was designed to aid the
selection of papers and piloted before use. A third reviewer was
designated to resolve any disagreement over the inclusion of any
particular trial.
Data extraction
We designed and piloted a data extraction form before use on
the selected papers. Data extraction was duplicated by the second
reviewer (NP) and any disagreements resolved by the third reviewer
(KS). Wherever possible the following data were extracted from
the selected studies:
1. Details of participants including demographic characteristics,
source of recruitment, site of oedema, cause of oedema, duration
of oedema, relevant co-existing medical conditions, and details of
the subjects’ occupations.
2. Where relevant, details of the type and treatment of any cancer,
including the patient’s cancer status at the time of the trial.
3. Details of the experimental and control interventions, including
the length of time they were applied for and confirmation that
CDT had not been used in the 6 months before the trial.
4. Details relating to the homogeneity of the two treatment groups,
e.g. the severity of oedema at the start of the trial and/or the
duration of oedema.
5. The methods of assessment of limb volume and other relevant
outcomes.
6. Details of any financial support that might introduce a conflict
of interest;
7. The numbers of participants allocated to each group and the
numbers lost to follow-up or excluded, together with the reasons
why.
Analysis
5Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
None of the studies identified compared the same intervention,
therefore data were not combined. We did not perform sub-group
analysis due to insufficient data. Analysis was by intention to treat.
As no studies were found comparing the same interventions, the
results of each study have been described in text form. Further
statistical analysis was not possible.
There were insufficient data to retrieve information on infection
and quality of life.
R E S U L T S
Description of studies
See: Characteristics of included studies; Characteristics of excluded
studies; Characteristics of ongoing studies.
Search results
As a result of the 2003 search, we identified 195 papers for possible
inclusion. Of these, 185 were found to be ineligible on the basis
that they were not randomised controlled trials. Of the remain-
ing 10 studies, all were described as randomised. For one study
(Thiadens 1999) only the abstract was available and to date the
review authors have been unable to obtain the full report; the de-
sign of a trial reported in another paper (Bergan 1998) remains
uncertain and awaits clarification. These are listed in the Studies
Awaiting Assessment Table. Of the remaining eight studies, data
were only extracted from only three as the other five did not meet
the inclusion criteria. Reasons for their exclusion are summarised
in the Characteristics of Excluded Studies Table.
The 2006 search of the specialised register identified a further
possible 33 studies and the 2007 search identified a further possible
21 studies. Of these, eight studies were retrieved as potentially
eligible RCTs and read in full by two reviewers (AL and SE). Two
studies are ongoing (Didem 2008 (this is longer follow up of one
of the publications excluded because of short follow up); Schmitz
2006) and are listed in the Ongoing Studies Table. Seven studies
did not fulfill the eligibility criteria for the review and are listed in
the Excluded Studies Table. The 2008 search of other electronic
databases identified a further possible 477 studies. One study was
retrieved for further inspection but was excluded as it did not meet
the eligibility criteria (Irdesel 2007).
Most of the potentially relevant studies were excluded because fol-
low up was less than 6 months. These short term trials included 2
studies of pneumatic compression (Dini 1998; Johansson 1998),
five studies comparing MLD with various types of control (Sitzia
2002; Williams 2002; Didem 2005; Wilburn 2006; McNeely
2004), one study of compression plus exercise vs exercise alone (
Johansson 2005) and one study of upper body exercise (McKenzie
2003). Another study had six month follow up but most partic-
ipants continued with the experimental intervention, pneumatic
compression, after 2 months follow up (Szuba 2002). Two studies
were excluded because limb size was not assessed using the volume
measurement (Bertelli 1991; Irdesel 2007) and one study was ex-
cluded because only a proportion of the participants (29%) had
lymphedema at baseline (Ahmed 2006). Full details of all of the
excluded studies are found in Table 1.
Table 1. Details of excluded trials
Study ID Methods Participants Interventions Outcomes Notes
Ahmed 2006
USA
6 month parallel
group trial
Blocked randomisa-
tion procedure with
random number ta-
ble and stratifying
according to age and
body fat
2 groups:
(1) weight training
twice a week (n=23)
(2) control
- no weight training
(n=23)
Recruitment
was between Octo-
ber 2001 and June
2002.
Breast cancer sur-
vivors living in the
greater Minneapo-
lis-Saint Paul area.
All appear to have
had axillary node
dissection.
Only a proportion
of the group had
lymphoedema
(1) Weight training
twice a week for 6
months for approx 1
hour
(2) Control - not
clearly described. It
appears they had no
weight training.
- Change in arm cir-
cumference
- Incidence
of lymphoedema at
6 months (measured
by arm circumfer-
ence or self report)
Only a proportion
of the participants
had lymphoedema
at baseline (13%).
Trial does not use
limb volume as the
method of assessing
change in size.
6Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Details of excluded trials (Continued)
Withdrawn/
excluded:
Not clear. It appears
as though the anal-
yses were only un-
dertaken on women
who had axillary
dissection but the
numbers vary from
22 to 23 in control
group. Dropouts
from original trial:
(1) 4/42 (10%) -
2 breast cancer re-
currence, 2 personal
reasons; (2) 3/43
7% - 2 breast cancer
recurrence, 1 per-
sonal reasons
(measured by arm
circumference mea-
surements, self re-
port of diagnosis
and self report of
symptoms)
(1)
Age: 52.3 (7.7) yrs
Self report of lymph
diagnosis: 7
(30.4%)
Self report of lymph
symptoms: 10
(43.4%)
Lymph by arm circ:
4 (17.4%)
Time since br ca di-
agnosis: 22.3 (7 to
43) mths
(2)
Age: 51.7 (7.5) yrs
Self report of lymph
diagnosis: 6
(26.1%)
Self report of
lymph symptoms: 7
(30.4%)
Lymph by arm circ:
4 (17.4%)
Time since br ca di-
agnosis: 21.9 (11 to
57) mths
Bertelli 1991
Italy
6 month parallel
group trial
Method of
randomisation not
stated.
Two groups:
Group 1: Hosiery
alone (n=37)
Group 2: Hosiery +
electrically
stimulated drainage
(ESD) (n=37)
All had unilateral
oedema of the up-
per limb following
trt for breast ca.
All had axillary node
dissection.
All had Delta value
>10 cms <20 cms
Group 1:
Median (range)
Age = 64 (45-77)
Group 1:
Compression sleeve
worn for 6 hours a
day
Group 2:
Compression sleeve
worn for 6 hours a
day
ESD 10x30 min
over 2 weeks; gap of
5 weeks then same
cycle repeated.
- Change in size es-
timated by taking
circumference mea-
sure-
ments at 7 points
along both swollen
& normal limb, es-
tablishing the differ-
ence between them,
and totalling these.
the sum of the dif-
ferences be-
tween swollen and
normal were desig-
Trial is long enough
but does not use
limb volume as the
method of assessing
change in size.
7Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Details of excluded trials (Continued)
Withdrawn/
excluded:
14/74
Group 1: 6/37
2 refused treatment
4 lost to follow-up
Group 2: 8/37
2 refused treatment
1 had lymphangitis;
2 lost to follow-up;
3 withdrawn be-
cause oedema wors-
ened.
No. (%)
Radical mast = 17
(46)
Mod. radical mast =
12 (32.4)
Quandrantectomy
= 8 (21.6)
Radiotherapy = 14
(37.8)
Chemotherapy = 10
(27)
Prev.
trt for l’oedema = 19
(51.3)
Prev. lymphangitis =
12 (32.4)
Group 2:
Median (range)
age = 64 (48-78)
No. (%)
Radical mast = 10
(27)
Mod. radical mast =
18 (48.7)
Quandrantectomy
= 9 (24.3)
Radiotherapy = 14
(37.8)
Chemotherapy = 6
(16.2)
Prev.
trt for l’oedema = 16
(43.2)
Prev. lymphangitis =
12 (32.4)
Both groups re-
ceived advice of skin
care and prevention
of infection.
nated “Delta”.
Response was deter-
mined by the %
change in the fi-
nal Delta value com-
pared to baseline.
Didem 2005
Turkey
4 week trial.
Randomi-
sation by unmarked
envelopes.
2 groups:
(1) Complex decon-
gestive physiother-
apy (CDP) (n=27)
(2) Standard phys-
iotherapy (SP)
All had developed
lymphoedema after
the first year from
surgery (average du-
ration of lymph at
time of treatment
was 3 years).
60% had moder-
ate lymph; 40% had
mild
lymph (but this var-
Group 1:
CDP: lymph
drainage,
multilayer compres-
sion bandage, eleva-
tion, remedial exer-
cises and skin care
Group 2:
SP: bandage, eleva-
tion, head-neck and
shoulder exercises
- Edema of the arm
(assessed by circum-
ference
measurements)
- Volume of the arm
(assessed with water
displacement)
- Shoulder mobility
(assessed by exten-
sion-flexion, abduc-
tion-adduction, ex-
ternal rotation)
Trial does not fol-
low the patients for
long enough. The
authors state that
they plan to report
on long term fol-
low up (6 months,
12 months and 24
months). The au-
thors were
contacted but have
not submitted their
8Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Details of excluded trials (Continued)
(n=26)
Withdrawn/
excluded:
There appear to be
no withdrawals af-
ter randomisation.
Prior to randomi-
sation: 1/28 from
group 1 withdrew
because of br ca re-
currence; 2/28 with-
drew from group 2
because of arm in-
fection.
ied by randomised
group (there are er-
rors in the publica-
tion).
All patients had un-
dergone axillary dis-
section with a range
of 2 - 35 nodes re-
moved.
(1)
Mild lymph: 7.4%
Moderate lymph:
92.3%
Mild: age 53.1 (3.1)
yrs
Moderate: age 61.3
(7.2) yrs
Mild: weight 66.3
(9.8) kg
Moderate: weight
73.9 (13.2) kg
Mild: history of cel-
lulites 8.3%
Moderate: history
of cellulites 20%
(2)
Mild lymph: 73.1%
Moderate lymph:
26.9%
Mild: age 54.7
(12.1) yrs
Moderate: age 63.6
(0.7) yrs
Mild: weight 64.7
(11.4) kg
Moderate: weight
71.5 (6.8) kg
Mild: history of cel-
lulites 0%
Moderate: history
of cellulites 29.4%
and skin care
All patients had
their physiotherapy
once a day, 3 days
a week for 4 weeks
and they were blind
to the intervention.
longer term follow
up.
Dini 1998
Italy
9 week parallel
group trial
Randomisation
achieved by phone
All parti-
ciants had unilateral
upper limb oedema
following “radical”
breast surgery + axil-
Group 1:
Pneu-
matic compression
(60mmHg) 2hr ses-
- Change in size es-
timated by taking
circumference mea-
sure-
Trial does not follow
patients up for long
enough (i.e. at least
6 months).
9Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Details of excluded trials (Continued)
call to Clinical trials
office
Two groups
Group 1:
Pneumatic
Compression Ther-
apy [PCT] (n=40)
Group 2:
No PCT (n=40)
Withdrawn/
excluded:
13 in all
Group 1 no. 8
2/40 recur ca.
3/40 refused PCT
3/40 lost to follow-
up
Group 2 no. 5
2/40 recurrence of
cancer
3/40 lost to follow-
up
lary node dissection
for stage I-III breast
ca.
All had “Delta”
value of > 10cms.
Group 1:
Mean (+/-)
Age = 62 (12)
duration of oed.
(days) = 155 (38)
Baseline Delta =
16.6 (6.7)
No. (%)
Radical Mast = 4
(10)
Mod. radical Mast.
= 20 (50)
Conservative surg =
16 (40)
Radiotherapy = 21
(52.5)
Adj. Chemo = 11
(27.5)
Prev. lymphangitis =
6 (15)
Group 2:
Mean (+/-)
Age = 62 (10)
duration of oed.
(days) = 236 (59)
Baseline Delta =
14.9 (4.6)
No. (%)
Radical Mast = 9
(22.5)
Mod. radical Mast.
= 23 (57.5)
Conservative surg =
8 (20)
Radiotherapy = 14
(35)
sions 5 days a week
for 2 wks; gap of 5
weeks then repeat of
same schedule as be-
fore.
Group 2:
No treatment
Both
groups received ad-
vice on care of skin
& prevention of in-
fection.
No other physical
treatments were al-
lowed.
ments at 7 points
along both swollen
& normal limb, es-
tablishing the differ-
ence between them,
and totalling these.
the sum of the dif-
ferences be-
tween swollen and
normal were desig-
nated “Delta”.
Response was de-
termined by the %
change in the fi-
nal Delta value com-
pared to baseline.
Also, does not use
limb volume as the
method of assessing
change in size
10Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Details of excluded trials (Continued)
Adj. Chemo = 8
(20)
Prev. lymphangitis =
10 (25)
Irdesel 2007
Turkey
6 month parallel
group trial
Ran-
domisation method
not clear from ab-
stract
2 groups:
(1) ex-
ercises and compres-
sion garment, n=10
(2) exercises alone,
n=9
Withdrawn/
excluded: Not clear
from abstract
Details not clear
from abstract
Group 1:
Exercises and com-
pression garment
Group 2:
Exercises alone
- Reduction in lym-
phoedema volume
(assessed by mea-
surement of arm cir-
cumference)
- Improvement in
shoulder range of
motion
- Symptoms such
as pain and tender
points
Trial does not use
limb volume as the
method of assessing
change in size
Johansson 1998
Sweden
4
week parallel group
trial in 2 parts
Part I all subjects
received same treat-
ment
Part II randomised
to 2 different inter-
ventions
Method of
randomisation not
stated
2 groups
Group 1:
Manual Lymph
Drainage [MLD]
(n=14)
Group 2:
Sequential Pneu-
matic Compression
[SPC] (n=14)
Withdrawn/
Recruitment
was over a 2.5 year
period.
Patients attend-
ing Dept. of phys-
ical therapy with
unilateral breast ca-
related upper limb
lymphoedema.
All had axillary node
dissection.
Lymphoedema de-
fined as excess limb
vol >10%
Group 1:
Median (q1-q3)
Age = 64 (52.5-
69.5)
Duration of oedema
(months) 14.0 (3.0-
76.5)
All participants be-
gan with 2 weeks of
wearing a compres-
sion sleeve
3rd week
Group 1:
Received MLD for
45 mins/day; 5/7 for
2 weeks at same time
each day
Group 2 :
Received sequen-
tial pneumatic com-
pression therapy
with the Lympha-
Press 2hrs/day (40-
60 mmHg)
Both groups contin-
ued to wear a com-
pression sleeve.
- % reduction in ex-
cess limb volume
measured by water
displacement
- Body weight
- Passive mobility of
shoulder measured
by goniometer
- Isometric muscle
strength measured
by dynamometer
- Patient’s subjective
assessment of: pain,
heaviness, tension,
function, paresthe-
sia.
Trial does not follow
patients up for long
enough (i.e. at least
6 months)
11Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Details of excluded trials (Continued)
excluded
End of Part 1 - 1/28
because oedema re-
solved
Part 2 - 4 in all, 2
from each group.
2 patients had recur-
rent cancer;
1 patient had
erysipelas;
1 patient could not
attend for follow-up
No.
Partial mastectomy
= 1
Mastectomy = 11
Radiotherapy = 10
Group 2:
Median (q1-q3)
Age = 57.5 (47.5-
69.5)
Duration of oedema
(months) 6.0 (2.3-
68.3)
No.
Partial mastectomy
= 2
Mastectomy = 10
Radiotherapy = 8
Johansson 2005
Sweden
5 day crossover trial.
Randomisation
method not stated.
2 groups:
(1)
arm exercises with
compression sleeve
(n=16 first)
(2) arm exercises
without
compression sleeve
(n=15 first)
Withdrawn/
excluded:
4/42 withdrew be-
cause of personal
reasons; 7/42 did
not reach the eligi-
bility criteria of at
least a 10% greater
arm volume in the
affected arm (after
randomisation)
Recruitment was in
September 2002.
Patients from the
physiothera-
pists registry of lym-
phoedema patients
at the Physiother-
apy Department of
Vaxjo Central Hos-
pital and the Lym-
phedema
Unit, Lund Univer-
sity Hospital, Lund
Sweden.
Lymphoedema de-
fined as 10-40%
greater arm volume
in affected arm.
Demographic char-
acteristics not given
separately by group.
Age: 55.3 (7.3) yrs
Surgery:
partial n=13; mas-
All participants
wore a compression
sleeve for 2 weeks
before the trial.
Trial was over 5
days:
Training exercise on
day 1, assessments
before, immediately
after and 24 hours
after.
Then training exer-
cise on day 3 with
assessments before,
imme-
diately after and 24
hours after. The par-
ticipants appear to
be crossed over to
the alternate treat-
ment for the 2nd
training session.
- Arm volume (mea-
sured by water dis-
placement method
and multiple fre-
quency bioelectrical
impedance analysis)
- Borg’s scale for per-
ceived exertion
Trial does not follow
patients up for long
enough (at least 6
months)
12Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Details of excluded trials (Continued)
tectomy n=18
No of axillary nodes
dissected: 13.3
(5.2)
Duration of edema:
66.7 (51.7) months
BMI: 25.9 (3.2)
McKenzie 2003
Canada
8 week parallel
group trial.
Randomisation
method not stated.
2 groups:
(1) Exercise, n=7
(2) Control (no ex-
ercise), n=7
Withdrawn/
excluded:
There did not ap-
pear to be any with-
drawals from the
trial.
Recruitment details
not provided.
Lymphoedema de-
fined
as unilateral and be-
tween 2 and 8 cm on
at least one measure-
ment point.
Group 1:
Age: 56.4 (10.4) yrs
Weight: 77.8 (20.6)
kg
BMI: 29.1 (6.6)
Group 2:
Age: 56.9 (8.2) yrs
Weight: 67.3 (9.1)
kg
BMI; 25.6 (3.3)
A compression
sleeve was used daily
by all participants.
(1) 8 week exer-
cise program (re-
sistance training 3
times per week plus
aerobic exercise us-
ing a Monark Re-
hab Trainer arm er-
gometer)
(2) No specific exer-
cise instruction
- Arm circumfer-
ence
- Arm volume (as-
sessed by water dis-
placement)
- SF36
Trial does not fol-
low patients for long
enough (at least 6
months).
McNeely 2004
Canada
4 week parallel
group trial.
Randomisation by
computer generated
code.
2 groups:
(1) MLD plus
multi-layered com-
pression bandaging,
n=25
(2) Com-
pression bandaging
alone, n=25
Withdrawn/
excluded:
5/50 did not com-
Recruit-
ment from Novem-
ber 2000 to De-
cember 2001 from
Cross Cancer Insti-
tute in Edmonton,
Canada.
Lymphoedema de-
fined as a mini-
mum of a 150ml
difference between
affected and unaf-
fected arms.
Group 1:
Age: 58 (13) yrs
No lymph nodes re-
moved: 12 (6)
All participants had
standard education
on proper arm and
skin care.
(1) 45 minutes of
daily MLD 5 times
per week (Vodder
method) plus short
stretch bandaging
(2) Short stretch
bandaging alone
- Reduction in arm
lymphoedema vol-
ume (assessed by
water displacement
volumetry and arm
circumference)
This outcome was
assessed by indepen-
dent asses-
sors blinded to sub-
ject treatment as-
signment.
Trial does not fol-
low patients for long
enough (at least 6
months).
13Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Details of excluded trials (Continued)
plete the study. 2
withdrew because of
adverse events (1
from group 1 be-
cause of a skin reac-
tion to bandaging,
1 from group 2 be-
cause of elbow dis-
comfort from
bandaging), 3 from
group 2 withdrew (1
because of personal
reasons, 2 because of
dissatisfaction with
treatment response)
. 1 additional partic-
ipant from group 1
withdrew from the
assessment of arm
volume because of
an error in measure-
ment.
Type of lymph: mild
31%, moderate
52%, severe 17%
Early lymp 35%
Chronic lymph
65%
Group 2:
Age: 63 (13) yrs
No lymph nodes re-
moved: 10 (5)
Type of lymp: mild
24%, moderate
52%, severe 24%
Early lymph: 48%
Chronic lymph:
52%
Sitzia 2002
UK
2 week parallel
group trial .
Randomisation
by computer-gener-
ated code managed
by non-clinical re-
searcher.
Two groups
Group 1: MLD
(n=15);
Group 2: SLD
(n=13).
Withdrawn/
excluded:
1 pt from Group 2
only treated for 5
days.
No adverse events
Re-
cruited from all new
referrals to l’oedema
clinic from Jan ’96
to June ’99.
All were
female, with unilat-
eral upper
limb oedema fol-
lowing trt for breast
ca.
95 referred;
40 fulfilled inclu-
sion criteria;
12 declined;
28 consented.
Group 1
Mean (SD):
age = 68 (10.8)
% Excess Volume at
start = 68.3 (39.5)
No. (%)
Mastectomy & ax-
Group 1:
MLD given for 40-
80 mins 5 days a
week for 2 weeks;
Group 2:
SLD performed by
same therapist 20
mins 5 days a week
for 2 weeks
Following mas-
sage limbs in both
groups were ban-
daged using multi-
layer technique
- % Reduction in ex-
cess limb volume at
2 weeks.
- Limb volume cal-
culated from surface
measurements.
Trial does not follow
patients up for long
enough (i.e. for at
least 6 months)
14Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Details of excluded trials (Continued)
illary clearance =
11(73)
Lumpectomy & ax-
illary clearance = 2
(13)
Fine needle aspira-
tion = 1 (7)
No surgery = 1 (7)
Radiotherapy = 13
(87)
Group 2
Mean (SD):
age = 75 (10.2)
% Excess Volume at
start = 58.5 (26.7)
No. (%)
Mastectomy & ax-
illary clearance =
11(85)
Lumpectomy & ax-
illary clearance = 2
(15)
Fine needle aspira-
tion = 0
No surgery = 0
Radiotherapy = 10
(77)
Szuba 2002
USA
6 month crossover
trial.
Randomisation
method not stated.
First phase of study:
10 day RCT with 30
day follow up of IPC
+ DLT vs DLT.
Sec-
ond phase of study:
2 month crossover
RCT with 6 month
follow up of IPC +
DLT vs DLT.
2 groups:
(1) Inter-
mittent pneumatic
Re-
cruitment from pa-
tients presenting to
Stanford Centre for
Lymphatic and Ve-
nous Disorders with
stable treated lym-
phoedema.
Lymphoedema was
defined as presence
of an increase of at
least 20% in the vol-
ume of the swollen
limb compared to
the volume of the
contralateral arm.
Demographic
data not reported by
All participants had
completed an initial
course of intensive
DLT btwn 1 and 12
months prior to en-
rolment.
(1) IPC, 1 hr daily
administered
at home + daily DLT
(MLD for 30 to
60 minutes (Vodder
type), compressive
bandaging and de-
congestive exercises
(2) Daily DLT alone
- Arm volume (as-
sessed by water dis-
placement method)
- Tissue tonometry
(to assess elasticity
of the skin)
- Goniometry (to
measure joint mo-
bility)
All participants were
free to con-
tinue with IPC after
2 months of treat-
ment, so assessment
at 6 month follow
up was not of ran-
domised groups.
15Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Details of excluded trials (Continued)
compression (IPC)
+ decongestive lym-
phatic
therapy (DLT) first,
DLT alone second,
n=13
(2) DLT alone first,
IPC + DLT second,
n=12
(Each participant
had both treatments
in a random order)
Excluded/
withdrew:
2/27 did not com-
plete the study. After
the 2 month RCT,
20/25 continued us-
ing IPC and 19 were
still using IPC at
end of follow up.
group.
All participants:
Age: 65.9 (43 to 81)
yrs
Average duration of
lymph: 60 (3 to
480) mths
Average time from
surgery: 113.7 mths
Wilburn 2006
USA
6 week crossover
trial.
Randomisation
method not stated.
2 groups:
(1) Flexitouch (me-
chanical de-
vice which simulates
MLD) first, massage
second, n=5
(2) Massage first,
Flexitouch second,
n=5
All participants used
a compression gar-
ment daily
Withdrawn/
excluded:
There do not ap-
pear to be any with-
drawals.
Re-
cruitment from pa-
tients presenting to
Stanford Centre for
Lymphatic and Ve-
nous Disorders.
Lymphoedema de-
fined as an increase
of at least 10% in
the mea-
sured volume of the
affected arm when
compared with the
contralateral limb.
Demographic
data not reported by
group.
All participants:
Age: 60 (7) yrs
Duration since ini-
tial cancer Rx: 103
(87) mths
On-
set of arm swelling
(1) Flexitouch self
administered for 1
hour daily for 14
consecutive days
(2) Self adminis-
tered massage for 1
hour daily for 14
consecutive days
1 week washout pe-
riod between treat-
ments where the use
of the compression
garment only was
permitted.
- Limb volume (as-
sessed by surface
measurements and
a simplified formula
for a ftrustum)
- SF36
Trial does not follow
patients up for long
enough (i.e. for at
least 6 months).
16Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Details of excluded trials (Continued)
prior to enrolment:
34 (34) mths
Weight: 75 (12) kg
Williams 2002
UK
9 week crossover
trial.
Method of
randomisation not
stated
Two trial groups
Group 1: MLD first
self-administered
massage [SAM] sec-
ond (no=15)
Group 2: SAM first,
MLD second
(n=16)
Withdrawn/
excluded:
2/31 both from
Group 2, 1 had pul-
monary embolus, 1
herpes zoster in af-
fected area.
All patients had uni-
lateral upper limb
lymphoedema as a
result of trt for
breast ca.
All had excess limb
vol > 10%, 2 consec-
u-
tive stable arm mea-
surements, evidence
of trunk oedema.
Group 1:
Mean (SD calcu-
lated from SE)
Age = 59.7 (8.1)
Baseline % Ex Vol =
30.1 (18.9)
Duration of oedema
(months) = 82.5
(56.9)
Weight (kg) = 70.8
(12.0)
No. (%)
Local excision = 9
(60)
Mastectomy = 6
(40)
Axillary sampling =
8 (53)
Axillary clearance =
6 (40)
no surgery = 1 (7%)
Radiotherapy to
breast = 12 (80)
Radiotherapy to ax-
illa = 10 (67)
Chemo = 3 (20)
Tamoxifen = (7 (47)
Group 2:
Mean (SD calcu-
Group 1:
15 x 1 hour ses-
sions of MLD over
3 weeks, performed
by therapists fol-
lowed by 6 weeks
washout period of
no trt, then 3 weeks
of 20 mins daily
SAM.
Group 2:
3 weeks of 20 mins
daily
SAM. followed by 6
weeks washout pe-
riod of no trt, then
15 x 1 hour ses-
sions of MLD over
3 weeks, performed
by therapists.
Both
groups wore com-
pression sleeves dur-
ing the day through-
out the study.
Changes in excess
limb volume. Vol-
ume measured elec-
tronically by Perom-
eter and also cal-
culated from sur-
face measurements
taken by hand.
Changes in caliper
creep on right and
left axillary folds,
measured by modi-
fied Harpenden skin
calipers.
Changes in dermal
depth on affected
side, measured by
skin ultrasound.
Changes in quality
of life measured by
EORTC QLQ C30
Changes in move-
ment & function
using validated as-
sessment tool.
Subjective as-
sessment of changes
in altered sensation.
Change in arm and
trunk swelling
assessed subjectively
by “pinch test”.
Sub-
jective assessment of
change in areas of fi-
brosis.
Trial does not follow
patients up for long
enough (i.e. for at
least 6 months)
17Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Details of excluded trials (Continued)
lated from SE)
Age = 59.3 (9.6)
Baseline % Ex Vol =
39.5 (17.6)
Duration of oedema
(months) = 118.4
(88)
Weight (kg) = 72.7
(13.6)
No. (%)
Local excision = 8
(50)
Mastectomy = 8
(50)
Axillary sampling =
12 (75)
Axillary clearance =
4 (25)
no surgery = 0
Radiotherapy to
breast = 14 (88)
Radiotherapy to ax-
illa = 11 (69)
Chemo = 6 (38)
Tamoxifen = 7 (19)
Record of infection.
Patients perceptions
of treatment based
on diaries and inter-
views.
Adj. Chemo:
EORTC QLQ
C30:
CDP:
DLT:
ESD:
IPC:
L’oedema:
Mast:
Mins:
Oed:
Plt:
Adjuvant
chemotherapy
European organisa-
tion for Research
and
Treatment of Can-
cer Core 30 Qual-
ity of Life Question-
naire
Complex deconges-
tive physiotherapy
Decongestive lym-
phatic therapy
Electrically stimu-
lated drainage
Intermittent pneu-
matic compression
18Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Details of excluded trials (Continued)
q:
SAM:
SD:
SE:
SLD:
SP
Surg:
Trt:
V:
Lymphoedema
Mastectomy
Minutes
Oedema
P
Quartile
Self administered
massage
Standard deviation
Standard error
Simple lymph
drainage
Standard
physiotherapy
Surgery
Treatment
Volume
Details of the included studies
The three eligible studies included a total of 150 participants. A
summary of the participants, interventions and outcomes can be
found in the Table of Included Studies and are also detailed here.
Manual Lymph drainage study (MLD)
Andersen 2000 designed a 12 month parallel group trial, with
crossover from the control group to the MLD group after three
months, if participants felt their response was unsatisfactory.
All the participants were women with unilateral lymphoedema
of the upper limb following treatment for breast cancer. Patients
with recurrent cancer or who were receiving treatment for lym-
phoedema in the three months preceding the trial were excluded.
Women with an excess limb volume of more than 30% were ex-
cluded and offered intensive treatment with bandages but, rather
confusingly, those who did not want the intensive treatment were
allowed back into the trial. The investigators defined the mini-
mum level of oedema for inclusion not as the percentage excess
volume but as an absolute excess volume of more than 200mls.
Participants were randomised to MLD followed by self-adminis-
tered massage or no massage at all. The MLD group received eight
sessions of MLD in two weeks and training in self-massage to be
done daily for the remainder of the 12 month trial. Both control
and MLD groups wore flat-knit compression sleeves daily, were
19Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
instructed in exercises aimed at enhancing lymph flow and in skin
care and prevention of infection.
The main outcome was a change in the percentage excess limb
volume. Volume was calculated from surface measurements using
“piecewise quadratic approximation, known as Simpson’s rule of
integration”. The authors decided on a very complex approach
to analysis of the data on volume. The investigators also assessed
shoulder mobility on two planes: extension-flexion and adduc-
tion-abduction and asked patients to complete a questionnaire on
symptoms related to lymphoedema and the European Organisa-
tion for Research and Treatment of Cancer (EORTC QLQ C30)
questionnaire.
Hosiery Study
Hornsby 1995 investigated the use of hosiery on its own, rather
than in conjunction with another therapy such as MLD.
The study was described as a randomised trial but the method of
randomisation was not stated nor was the length of the trial. It
was of a parallel group design with two study groups.
Participants were recruited from patients attending a local follow-
up breast clinic, who, if they complained of swelling in the arm,
were referred on to a lymphoedema clinic set up especially for the
trial.
Both groups were taught exercises, skin care and self-administered
massage. The experimental group were fitted with elastic compres-
sion sleeves that were worn day and night.
The description of the method for assessing the main outcome,
limb volume, is a little confusing. It would appear that limb volume
was measured by water displacement in two stages, level with a
mark 10 cms below the olecranon and then level with a mark
15 cms above the olecranon. The normal limb volume at 15 cms
above the olecranon was subtracted from that of the swollen limb
and recorded at each visit. It is not clear why two marks were
needed or whether the final measurement was a combined total or
was minus the volume up to the level of the first mark. At each visit
participants were asked how much their swollen limb interfered
with daily activities and whether they had any pain.
Bandage Study
Badger 2000 investigated multi-layer bandaging. The trial took
place in London and involved two of the authors of this review.
The trial was described as randomised and the method of ran-
domisation was reported. There were two trial groups running in
parallel and the trial lasted six months.
Participants were recruited from the patients attending the lym-
phoedema clinic and included those with unilateral lymphoedema
of upper or lower limb, primary in origin as well as lymphoedema
secondary to the treatment of cancer.
Participants in the bandage group received a 19 day course of
multi-layer bandaging and were then fitted with hosiery which
was worn daily for the remainder of the trial. Participants in the
hosiery alone group wore hosiery daily from the start of the trial.
Both groups received advice on skin care, were taught exercises
and self-administered massage (SAM) which they were asked to
perform daily.
The main outcome was reduction in excess volume that was either
calculated from surface measurements or measured electronically
using a Perometer(r). Body weight was also noted. Instances of
infection and venous thrombosis were noted throughout the trial.
Risk of bias in included studies
Methodological details of the three included studies (Andersen
2000; Badger 2000; Hornsby 1995) can be found in the table
’Characteristics of Included Studies’. These studies were also as-
sessed for risk of bias (Higgins 2007) (see Risk of Bias tables for
each study).
Randomisation method
The method of randomisation was not described in 2 studies (
Andersen 2000; Hornsby 1995). In the other study, random allo-
cation was by a centralised telephone system (Badger 2000).
Allocation concealment
None of the included studies provided details of allocation con-
cealment.
Blinding
None of the included studies appeared to be blinded.
Incomplete data
All of the studies had withdrawals and loss to follow up after ran-
domisation, but in one study (Badger 2000), this proportion was
not considered great enough to cause major bias in the results. In
this study, of the 90 patients randomised, seven were excluded or
withdrawn, four from the bandage plus hosiery group and three
from the hosiery alone group. One participant declined treatment
with bandages, two did not complete the course of bandaging, one
developed a recurrence of cancer, one insisted on regular MLD and
two never attended for follow-up. In the Anderson study, 2/44, one
from each trial group, were withdrawn because they were found
to be ineligible. After three months, 10 participants in the control
group crossed over to the MLD group. A further five participants
were excluded: one died of a heart attack, two developed a recur-
rence of cancer, one withdrew due to her husband’s illness and one
participant withdrew due to depression. In the Hornsby study, no
explicit report was given of the withdrawals or exclusions from the
25 participants. It is clear from the data table included in the paper
that the number of patients in the control group had dropped by
about half by the third visit (6/11 remained) and by the fourth
only one participant remained. No reasons were provided for this
reduction in numbers. In the treatment group, follow up went on
for eight visits but there was a steady falling off of participants
from the third visit. Out of 14 participants at the start of the trial,
data are available for only three by the end; again no reasons are
20Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
put forward for this reduction in numbers.
Selective outcome reporting
None of the included studies provided details of their study pro-
tocols, so selective outcome reporting cannot be excluded.
Other bias
One study (Hornsby 1995) did not provide details of the compa-
rability of randomised groups at baseline. For the other 2 studies,
there was no evidence of a difference in prognostic factors at base-
line.
Effects of interventions
Manual Lymph Drainage Study (Andersen 2000)
Of the 42 participants included in the analysis 22 were randomised
to MLD + hosiery and 20 to hosiery alone. Other than endocrine
therapy, received by 10 participants in the hosiery alone group and
only two in the MLD group, the groups were evenly balanced in
other respects such as age, duration of oedema, at the start of the
trial.
The median excess volume at the start of the trial is reported as
the absolute volume rather than as a percentage so it is difficult
to judge how bad the participants’ oedema was. By the end of the
trial, the mean percentage reduction in excess limb volume was
60% (95% CI 43 to 78%) in the hosiery alone group as against
48% (95% CI 32 to 65%) in the MLD group. This difference
was not significant. No differences were found between the two
groups in the symptom scores.
The authors concluded that improvements seen in both groups
were attributable to the use of compression sleeves and that MLD
provided no extra benefit at any point during the trial.
Hosiery Study (Hornsby 1995)
There are problems with the reporting of this trial. From the text,
it seems that the data in the tables represent the difference between
swollen and normal limb volume at each visit. The raw data for
each participant are tabulated but no means, standard deviations
or confidence intervals are calculated. It is possible to work out the
percentage reduction in excess limb volumes for each participant
from these data. But the drop out rate is so high, particularly in
the control group, that little information is available beyond the
second month in the treated group and beyond the first month in
the control group. Data concerning pain and how much swelling
interfered with daily activities are not reported in any detail and
not according to trial group.
The author concluded that, although this was a small study, the
results suggest that wearing a compression sleeve is beneficial and
that the high dropout rate in the control groups may have reflected
the subjects’ lack of progress.
Bandage Study (Badger 2000)
Data were available for analysis on 83 participants, 34 in the ban-
dage + hosiery group and 49 in the hosiery alone group. Around
two thirds of participants had upper limb oedema and most were
female. There was a satisfactory balance between the groups in
terms of age, duration of oedema, site, gender, body weight and
excess volume at the start.
Both groups achieved a reduction in percentage of excess limb
volume but that achieved by the bandage + hosiery group was
consistently greater than that of the hosiery alone group. On day
19, marking the end of the course of bandaging, the bandage +
hosiery group achieved a mean reduction of 33.5% (SD 16.9)
as against 9.6% (SD 20.4) in the hosiery alone group. Progress
at week 12 was maintained at week 24 in both groups but that
achieved by the bandage + hosiery group (mean 32.6% SD 33.2)
was almost double that of the hosiery alone group (19.6% SD
28.5).
The authors concluded that in this mixed group of patients ban-
dage + hosiery resulted in a greater reduction in excess limb volume
than hosiery alone and this difference in reduction was maintained
long-term.
D I S C U S S I O N
Any oedema of the surface tissues that involves a fluid component
is likely to be influenced by the application of external pressure, as
clinical experience world-wide has demonstrated over many years.
From a physiological point of view, difficulties are likely to arise
if fluid is trapped in fatty tissues (since fat absorbs pressure) or in
fibrotic tissues where it becomes difficult for fluid to be displaced.
Both of these scenarios are common in lymphoedema but despite
these problems few therapists specialising in the management of
lymphoedema doubt that improvements can be obtained through
treatment. The question is not so much ’can lymphoedema be
treated?’ but rather ’what treatments reduce swelling, and the mor-
bidity associated with swelling, most effectively?’ In addition to
this question, we need to ask ’what treatments produce lasting
improvements?’ since short-term improvements satisfy no one.
It appears that at present there is no drug or surgery that will re-
duce chronic oedema and allow the reduction to be maintained.
Physical therapies remain the most commonly used treatments
for lymphoedema and are usually combined in a treatment pro-
gramme, since the general view is that no one treatment is likely to
be successful on its own. The difficulty lies in establishing which
of these physical treatments plays the most critical part in reducing
and controlling swelling and which, if any, can be safely left out
of the treatment programme.
This systematic review extends the findings of earlier non-
Cochrane systematic reviews of physical therapies for lym-
phoedema (Harris 2001; Kligman 2004; Moseley 2007). All re-
views were limited by the lack of adequately designed randomised
controlled trials. Most of the studies conducted so far in this field
are either designed poorly or are poorly reported. Most are too
small and provide too little follow-up to be of any use. There is
a tendency to concentrate on one section of the lymphoedema
21Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
population (i.e. breast cancer patients) when the growing body
of evidence on prevalence and incidence suggests that lower limb
oedema, either of primary origin or secondary to cancer and other
conditions, is also a significant problem.
Our objective in this review was to assess the impact of physical
treatment programmes on the volume, shape, condition and long-
term control of oedema in lymphoedematous limbs and in addi-
tion to establish the psycho-social benefits of physical treatment.
We were not able to answer any questions relating to the effect
of treatment on the incidence of infection. While it is accepted
that assessing the effect of treatment on the patients’ psycho-social
well-being presents considerable challenges there is a significant
body of literature on this subject, particularly in relation to other
chronic conditions such as diabetes and rheumatoid arthritis, that
could be drawn upon. The few studies included in this review offer
little to increase our understanding in this area.
A U T H O R S ’ C O N C L U S I O N S
Implications for practice
Of the three trials included here, only one studied lower as well
as upper limb oedema. All three trials have their limitations and
have yet to be replicated, so their results must be viewed with
caution. There is weak evidence to support the use of multi-layer
bandaging over hosiery alone.
Implications for research
There is a clear and pressing need for well-designed, randomised
trials of the whole range of physical therapies if the best approach
to managing lymphoedema is to be determined. Trials of complex
physical therapy programmes, while not easy to conduct, are nev-
ertheless possible.
A C K N O W L E D G E M E N T S
Our thanks go to the members of the editorial office of the
Cochrane Breast Cancer Group for assisting in all stages of the
review and especially to Sharon Parker (Review Group Coordina-
tor) and Nicole Holcroft and Joanne Talenta (Trials Search Coor-
dinator).
Thanks are also expressed to the study investigators who provided
additional information and were generally helpful.
We acknowledge the contribution of Anne Lethaby and Julie
Brown (Cochrane updating project) for the 2008 update.
R E F E R E N C E S
References to studies included in this review
Andersen 2000 {published data only}∗ Andersen L, Hojris I, Erlandsen M, Andersen J. Treatment of
breast-cancer-related lymphedema with or without manual
lymphatic drainage--a randomized study. Acta Oncologica 2000;39
(3):399–405.
Badger 2000 {published data only}∗ Badger CMA, Peacock JL, Mortimer PS. A randomized,
controlled, parallel-group clinical trial comparing multilayer
bandaging followed by hosiery versus hosiery alone in the treatment
of patients with lymphedema of the limb. Cancer 2000;88(12):
2832–7.
Hornsby 1995 {published data only}∗ Hornsby R. The use of compression to treat lymphoedema.
Professional Nurse 1995;11(2):127–8.
References to studies excluded from this review
Ahmed 2006 {published data only}
Ahmed RL, Thomas W, Yee D, Schmitz KH. Randomized
controlled trial of weight training and lymphedema in breast cancer
survivors. Journal of Clinical Oncology 2006;24(18):2765–2772.
Bertelli 1991 {published data only}
Bertelli G, Venturini M, Forno G, Macchiavello F, Dini D.
Conservative treatment of postmastectomy lymphedema: a
controlled randomised trial. Annals of Oncology 1991;2:575–8.
Didem 2005 {published data only}
Didem K, Ufuk YS, Serdar S, Zumre A. The comparison of two
different physiotherapy methods in treatment of lymphedema after
breast surgery. Breast Cancer Research and Treatment 2005;93(1):
49–54.
Dini 1998 {published data only}
Dini D, Del Mastro L, Gozza A, Lionetto R, Garrone O, Forno G,
et al.The role of pneumatic compression in the treatment of post-
mastectomy lymphedema. A randomized phase III study. Annals of
Oncology 1998;9:187–90.
Irdesel 2007 {published data only}
Irdesel J, Kahram an Celiktas S. Effectiveness of exercise and
compression garments in the treatment of breast cancer related
lymphedema. Turkiye Fiziksel Tip ve Rehabilitasyon Dergisi 2007;53
(1):16–21.
Johansson 1998 {published data only}
Johansson K, Lie E, Ekdahl C, Lindfeldt J. A randomized study
comparing manual lymph drainage with sequential pneumatic
22Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
compression for treatment of postoperative arm lymphedema.
Lymphology 1998;31(2):56–64.
Johansson 2005 {published data only}
Johansson K, Tibe K, Weibull A, Newton RC. Low intensity
resistance exercise for breast cancer patients with arm lymphedema
with or without compression sleeve. Lymphology 2005;38(4):
167–180.
McKenzie 2003 {published data only}
McKenzie DC, Kalda AL. Effect of upper extremity exercise on
secondary lymphedema in breast cancer patients: a pilot study.
Journal of Clinical Oncology 2003;21(3):463–466.
McNeely 2004 {published data only}
McNeely ML, Magee DJ, Lees AW, Bagnall KM, Haykowsky M,
Hanson J. The addition of manual lymph drainage to compression
therapy for breast cancer related lymphedema: a randomized
controlled trial. Breast Cancer Research and Treatment 2004;86(2):
95–106.
Sitzia 2002 {published data only}
Sitzia J, Sobrido L, Harlow W. Manual lymphatic drainage
compared with simple lymphatic drainage in the treatment of post-
mastectomy lymphoedema: a pilot randomised trial. Physiotherapy
2002;88(2):99–107.
Szuba 2002 {published data only}
Szuba A. Literature watch. The addition of manual lymph drainage
to compression therapy for breast cancer related lymphedema: a
randomized controlled trial. Lymphatic Research and Biology 2005;3
(1):36–41.∗ Szuba A, Achalu R, Rockson SG. Decongestive lymphatic therapy
for patients with breast carcinoma-associated lymphedema. Cancer
2002;95:2260–2267.
Wilburn 2006 {published data only}
Wilburn O, Wilburn P, Rockson SG. A pilot, prospective
evaluation of a novel alternative for maintenance therapy of breast
cancer associated lymphedema. BMC Cancer 2006;6:84.
Williams 2002 {published data only}
Williams AF, Vadgama A, Franks PJ, Mortimer PS. A randomized
controlled crossover study of MLD therapy in women with breast
cancer-related lymphoedema.. European Journal of Cancer Care
2002;11(4):254–61.
References to studies awaiting assessment
Bergan 1998 {published data only}
Bergan JJ, Sparks S, Angle N. A comparison of compression pumps
in the treatment of lymphedema.. Vascular Surgery 1998;32(5):
455–62.
Thiadens 1999 {published data only}
Thiadens S, Wright E. The effectiveness of compression sleeves in
the maintenance of postmastectomy lymphedema. JOBST?
Beiersdorf Website Pre 1999.
References to ongoing studies
Didem 2008 {unpublished data only}
Comparison of two different physiotherapy methods in treatment
of lymphedema after breast surgery. Ongoing study Participants
treated between June 2002 and May 2003.
Follow up at 6 months, 1, 2 and 3 years after treatment was
initiated..
Schmitz 2006 {unpublished data only}
PAL (Physical Activity and Lymphedema) trial. Ongoing study
Start date: October 2005.
Finish date: June 2008..
Additional references
Badger 1988
Badger CMA, Mortimer PS, Regnard CFB, Twycross RG. Pain in
the chronically swollen limb. Progress in Lymphology - XI. Vienna:
Elsevier Science Publishers, 1988:243–5.
Badger 1997
Badger C. Shape. Chapter in: A study of the efficacy of multilayer
bandaging and compression hosiery versus hosiery alone in the
management of lymphoedema of the limbs and their effects on the
swollen limb. Unpublished Thesis, London 1997.
Carroll 1992
Carroll D, Rose KE. Treatment leads to significant improvement:
effect of conservative treatment on pain in lymphoedema.
Professional Nurse 1992;8(1):32–6.
Erickson 2001
Erickson VS, Pearson ML, Ganz PA, Adams J, Kahn KL. Arm
edema in breast cancer patients. Journal of the National Cancer
Institute 2001;93(2):96–111.
Foldi 1985
Foldi E, Foldi M, Weissleder H. Conservative treatment of
lymphoedema of the limbs. Angiology 1985;36(3):171–80.
Harris 2001
Harris SR, Hugi MR, Olivotto IA, Levine M. Clinical practice
guidelines for the care and treatment of breast cancer: 11.
Lymphedema. Canadian Medical Association Journal 2001;164(2):
191–199.
Higgins 2007
Higgins JPT, Altman DG (eds). Assessing risk of bias in included
studies (Part of Chapter 8 - under consultation). Cochrane
Handbook of Systematic Reviews of Interventions.
Kligman 2004
Kligman L, Wong RKS, Johnston M, Laetsch NS. The treatment of
lymphedema related to breast cancer: a systematic review and
evidence summary. Support Care Cancer 2004;12:421–431.
Levick 1991
Levick JR. An Introduction to Cardiovascular Physiology. First
Edition. London: Butterworth, 1991.
Moffatt 2003
Moffatt CJ, Franks PJ, Doherty DC, Williams AF, BAdger C, Jeffs
E, Bosanquet N, Mortimer PS. Lymphoedema: an underestimated
health problem. Quarterly Journal of Medicine 2003;96(10):731–8.
Mortimer 1995
Mortimer PS. Managing Lymphoedema. Clinical and Experimental
Dermatology 1995;20:98–106.
23Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Moseley 2007
Moseley AL, Carati CJ, Piller NB. A systematic review of common
conservative therapies for arm lymphoedema secondary to breast
cancer treatment. Annals of Oncology 2007;8(4):639–646.
Petrek 1998
Petrek JA, Heelan MC. Incidence of breast carcinoma-related
lymphoedema. Cancer 1998;12 Supple:2776–81.
Sitzia 1997
Sitzia J, Stanton AWB, Badger C. A review of outcome indicators in
the treatment of chronic limb oedema. Clinical Rehabilitation
1997;11:181–91.
Tobin 1993
Tobin M, Lacey HJ, Meyer L, Mortimer PS. The psychological
morbidity of breast cancer related arm swelling. Cancer 1993;72
(11):3248–52.
References to other published versions of this review
Badger 2004
Badger C, Preston N, Seers K, Mortimer P. Physical therapies for
reducing and controlling lymphoedema of the limbs. Cochrane
Database of Systematic Reviews 2004, Issue Issue 4. [Art. No.:
CD003141. DOI: 10.1002/14651858.CD003141.pub2]∗ Indicates the major publication for the study
24Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Andersen 2000
Methods 12 month trial
Method of randomisation not reported
2 Trial Groups
Group 1: Standard treatment alone (no.23)
Group 2: Standard treatment + MLD (no.21)
Group 1 pts allowed to cross over to Group 2 after 3 months if not satisfied with response.
Withdrawn / excluded
no.7 out of total 44:
2 / 44 pts (1 in each group) found to be ineligible:
1 had recurrent cancer;
1 less than 4 months post surgery;
1/44 withdrew due to husband’s illness (? which group);
1 /44 did not return at 12 months due to depression;
1/44 died
According to group:
Group 1:
2/23 local recurrence
Group 2:
1/21 died
Participants Recruited breast ca. patients attending lymphoedema clinic
44 randomized
Group 1: n22
[Median (range)]
Age 56yrs (29-77)
ExVol 361mls (78-1184)
Duration of oedema 12mnths (4-126)
Group 2: n20
[Median (range)]
Age 53yrs (25-73)
ExVol 340mls (161-1297)
Duration of oedema 15mnths (5-183)
Breast Cancer treatment:
Group 1:
axillary dissection n11;
No RT n11;
RT including axilla n4;
RT not including axilla n7
Group 2:
axillary dissection n15;
No RT n8;
RT including axilla n2;
RT not including axilla 10
25Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Andersen 2000 (Continued)
Interventions Group 1
Standard treatment consisted of compression sleeve worn during day - used decreasing
sizes for first 2 weeks then fitted with made-to-measure sleeves;
advice on exercises and skin care.
Group 2
Standard treatment as above plus 8 sessions of MLD in 2 weeks; then taught SAM to be
used daily.
Outcomes Change in limb volume, calculated from surface measurements;
questionnaire on related symptoms and on compliance with treatment.
Assessments performed at start, 3, 6, & 9 months.
Further assessment at 4 months for any patients crossing over.
Notes Author contacted re method of randomisation - no response.
Risk of bias
Item Authors’ judgement Description
Adequate sequence generation? Unclear No description provided
Allocation concealment? Unclear No description provided
Blinding?
All outcomes
No Very unlikely for any outcomes
Incomplete outcome data addressed?
All outcomes
No Of a total of 44 participants, 2 were ini-
tially excluded because of ineligibility. Af-
ter 3 months, 10 participants crossed over
to the other treatment and a further 5 par-
ticipants were excluded.
Free of selective reporting? Unclear Not clear
Free of other bias? Yes No indication of any other likely bias
26Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Badger 2000
Methods 6 Month trial
Central telephone randomisation
Group 1:
Multi-Layer Bandaging [MLB] + Hosiery, (n38);
Group 2:
Hosiery alone, (n52).
90 patients randomised,
results reported on 83.
Withdrawn/excluded:
7 in all
Group 1:
1 declined treatment;
1 developed recurrent Ca.
1 only had 14 days bandaging;
1 had only 11days bandaging.
Group 2:
1 insisted on continuing MLD;
2 never attended for follow-up
Participants Patients attending the Lymphoedema Service with unilateral lymphoedema; cancer-
related and non-cancer-related, affecting upper or lower limb with > 20% excess volume.
Recruited over a 2 year period
Details of cancer treatment not reported
Group 1
Mean (SD)
Age = 57.3yrs (14.5)
Duration of oedema = 48 months (96)
Baseline % ExVol = 48.6% (25.6)
Weight = 73Kgs (14)
N (%)
Male = 3 ( 9)
Female = 31 (91)
Upper limb = 21 (62)
Lower limb = 13 (38)
Group 2
Mean (SD)
Age = 57.4 yrs (14.6)
Duration of oedema = 60 months (96)
Baseline %ExVol = 41.9 % (25.6)
Weight = 71Kgs (13)
N (%)
Male = 4 (8)
Female = 45 (92)
Upper limb = 33 (67)
Lower limb = 16 (33)
Interventions Group 1:
18 day course of daily MLB kept in place around the clock, followed by compression
hosiery worn during the day for remainder of trial
Group 2:
27Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Badger 2000 (Continued)
Compression hosiery alone, worn daily throughout the whole of trial
Both groups were asked to exercise daily and perform SAM daily and advised on daily
skin care.
Outcomes % Reduction / increase in excess limb volume calculated from surface measurements or
from electronic measurements (Perometer)
Change in body weight
Notes
Risk of bias
Item Authors’ judgement Description
Adequate sequence generation? Yes Method of randomisation was by cen-
tralised telephone system
Allocation concealment? Unclear No details reported
Blinding?
All outcomes
No Blinding highly unlikely
Incomplete outcome data addressed?
All outcomes
Yes Dropouts not likely to cause major bias
Free of selective reporting? Unclear No details provided
Free of other bias? Yes No other bias detected
Hornsby 1995
Methods Length of trial not stated - follow-up appeared to continue for 7 months.
Method of randomisation not stated
Two trial groups:
Group1: Sleeve (n14)
Group 2: No sleeve (n11)
Withdrawn / excluded
By the end of the trial only 3 patients remained, all in Group 1.
No reason given for the dropouts other than to say it was presumably due to lack of
progress.
No adverse events reported
Participants Patients with oedema attending F-Up breast clinic between Nov ’91 to Dec ’92 referred
to a lymphoedema clinic set up for trial.
n60 referred;
n58 attended;
25 women consented to study
Not possible to say if groups are balanced as no details provided on:
28Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Hornsby 1995 (Continued)
age;
duration of oedema;
breast cancer treatment;
severity of oedema at outset (not possible to calculate the % Excess Volume at Start from
data provided as normal limb volumes are not reported).
Interventions Both groups were taught exercises and self-administered massage and given advice on
skin care.
Group 1:
compression sleeves worn day and night;
Group 2:
no treatment.
Outcomes Reduction / increase in excess limb volume measured by water displacement every 4
weeks
Patients were questioned about how much l’oedema interfered with daily living activities
at each visit. No information provided concerning the tool used.
Also questioned about pain at each visit. No information provided on the scale that was
used
Notes
Risk of bias
Item Authors’ judgement Description
Adequate sequence generation? Unclear No details provided
Allocation concealment? Unclear No details provided
Blinding?
All outcomes
No Blinding highly unlikely
Incomplete outcome data addressed?
All outcomes
No Significant number of dropouts and no rea-
sons given. Out of 14 participants at the
start of the trial, data is available for only 3
participants at the end.
Free of selective reporting? Unclear No details provided
Free of other bias? Unclear No details provided of comparability of
groups at baseline
Ca: Cancer
Ex Vol: Excess Volume
MLD: Manual Lymph Drainage
N: Number
PCT: Pneumatic Compression Therapy
29Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
RT: Radiotherapy
SAM: Self Administered Massage
SD: Standard Deviation
Characteristics of excluded studies [ordered by study ID]
Ahmed 2006 RCT comparing twice a week weight training with no weight training in 45 women with breast cancer associated
lymphoedema. Only a proportion of the participants (29%) had lymphoedema at baseline - see Table 1 for details.
Bertelli 1991 RCT comparing hosiery with electrically stimulated drainage versus hosiery alone in breast cancer associated
lymphoedema. Did not use limb volume as the method for assessing change in size - see Table 1 for details.
Didem 2005 RCT with crossover design comparing a complex decongestive physiotherapy treatment (lymph drainage, multi
layer compression bandage, elevation, remedial exercises and skin care) with standard physiotherapy (bandage,
elevation, head-neck shoulder exercises and skin care) in 53 patients with breast cancer associated lymphoedema.
Follow up was at the end of the fourth week of treatment - did not satisfy eligibility criteria of at least 6 months.
The authors state that they are going to report long term follow up results in a further publication (see Ongoing
Studies).
Dini 1998 RCT comparing pneumatic compression therapy (PCT) versus no PCT in 80 patients. Only lasted 9 weeks and
did not use limb volume as the method for assessing change in size - see Table 1 for details
Irdesel 2007 RCT comparing exercise plus compression with exercise alone in 19 patients with breast cancer associated lym-
phoedema. The efficacy of treatment was assessed by measurement of the arm circumference rather than limb
volume - see Table 1 for details.
Johansson 1998 RCT comparing manual lymph drainage versus sequential pneumatic compression in 28 patients. Only lasted 4
weeks - see Table 1 for details.
Johansson 2005 RCT comparing low intensity exercise plus compression sleeve with low intensity exercise alone in 31 patients
with breast cancer related arm lymphoedema. Measurements were taken immediately after treatment and 24 hours
later - see Table 1 for details.
McKenzie 2003 RCT comparing upper body exercise with no exercise in 14 patients with previous breast cancer. Treatment lasted
for 8 weeks and follow up was every 2 weeks until the end of the study - see Table 1 for details.
McNeely 2004 RCT comparing manual lymph drainage plus compression with compression alone in 50 women with breast
cancer related lymphoedema. Treatment and follow up lasted for 4 weeks - see Table 1 for details.
Sitzia 2002 RCT comparing manual lymph drainage and simple lymph drainage in 28 patients. Only lasted 2 weeks - see
Table 1 for details.
Szuba 2002 RCT with crossover design comparing intermittent pneumatic compression plus decongestive lymphatic therapy
with decongestive lymphatic therapy alone (MLD, compressive wrapping of the limb and decongestive exercises)
in 27 patients with breast cancer associated lymphoedema. Follow up was for 6 months, but after 2 months of
treatment, almost all the participants elected to take the experimental treatment - see Table 1 for details.
Wilburn 2006 RCT with crossover design comparing a mechanical device designed to simulate MLD (Flexitouch)with massage
in 10 patients with breast cancer associated lymphoedema. Duration of trial was 2 weeks, 1 week washout and 2
more weeks with alternate treatment - see Table 1 for details.
30Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
(Continued)
Williams 2002 RCT crossover trial comparing manual lymph drainage versus self-administered massage in 31 patients. Only
lasted 9 weeks - see Table 1 for details.
MLD: Manual Lymph Drainage
RCT: Randomised Controlled Trial
Characteristics of ongoing studies [ordered by study ID]
Didem 2008
Trial name or title Comparison of two different physiotherapy methods in treatment of lymphedema after breast surgery
Methods
Participants n=53.
Patients had developed unilateral lyphedema (mild-moderate)of duration at least 1 year after breast cancer
treatment.
Interventions (1) Complex decongestive physiotherapy (manual lymph drainage, compression bandage, remedial exercises
and skin care.
(2) Standard physiotherapy (bandage, elevation, head-neck and shoulder exercises and skin care)
Outcomes (1) Range of motion (extension-flexion, abduction-adduction, external rotation)
(2) Circumferential measurement of arm
(3) Volumetric measurement of arm (by water displacement)
Starting date Participants treated between June 2002 and May 2003.
Follow up at 6 months, 1, 2 and 3 years after treatment was initiated.
Contact information [email protected]
Notes Longer follow up of patients given treatment for 4 weeks. Contact was made with the author who stated that
analysis was ongoing at longer follow up.
Schmitz 2006
Trial name or title PAL (Physical Activity and Lymphedema)trial
Methods
Participants n=288
Participants were 1-15 years after breast cancer diagnosis, currently free of cancer recruited in 2 strata:
(1) women with stable lymphedema (5-15 years post diagnosis)
(2) women without lymphedema (1-5 years post diagnosis).
31Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Schmitz 2006 (Continued)
Interventions (1) Exercise intervention of twice weekly strength training (13 weeks of supervised training (90 mins/session)
+ 39 weeks of unsupervised training (90 mins/session))
(2) Non exercising control group
Outcomes (1) arm circumference
(2) extra-cellular water in the arm
(3) volumetry
(4) function tests (range of motion, pain, grip strength)
Starting date Start date: October 2005.
Finish date: June 2008.
Contact information [email protected]
Notes Contact made with author. Results forthcoming after completion of trial.
32Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
D A T A A N D A N A L Y S E S
This review has no analyses.
A P P E N D I C E S
Appendix 1. Search strategy for CENTRAL (Issue 1, 2008)
1. Lymphoedema or lymphedema or lymphodema or elephantiasis
2. exercise/
3. physical therapy
4. bandage
5. hosiery or hose
6. compression
7. 5 or 6 or 7 or 8 or 9
8. 1 and 7
Appendix 2. Search strategy for CINAHL (Ovid) (1982 to February 2008)
1. Lymphedema/or elephantiasis/
2. Lymph?dema or elephantiasis
3. 1 or 2
4. Physical therapy/
5. Bandages and dressings/
6. Compression garments/
7. physical therapy
8. bandage
9. hosiery or hose
10. compression
11. 4 or 5 or 6 or 7 or 8 or 9 or 10
12. 3 and 11
Appendix 3. Search strategy for MEDLINE (Ovid) (1960 to February 2008)
1. Lymphedema/or elephantiasis/
2. Lymph?dema or elephantiasis
3. 1 or 2
4. Physical therapy modalities/
5. Bandages/
6. Intermittent pneumatic compression devices/
7. physical therapy
8. bandage
9. hosiery or hose
10. compression
11. 4 or 5 or 6 or 7 or 8 or 9 or 10
12. 3 and 11
33Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Appendix 4. Search strategy for EMBASE (Ovid) (1980 to February 2008)
1. Lymphedema/or elephantiasis/
2. Lymph?dema or elephantiasis
3. 1 or 2
4. Physiotherapy/
5. Bandage/
6. Kinesiotherapy/
7. Intermittent pneumatic compression device/
8. Compression therapy/
9. physical therapy
10. bandage
11. hosiery or hose
12. compression
13. 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12
14. 3 and 13
Appendix 5. Search strategy for National Research Register (NRR) (February 2008)
1. lymphedema or lymphoedema or lymphodema
Appendix 6. Search strategy British Nursing Index, British Library Index, UnCover, PASCAL(September 2003)
Lymph?edema OR elephantiasis
a) AND physical therapy
b) AND bandage* OR (compression bandage*)
c) AND hosiery OR hose
d) OR hosiery OR hose NEAR compression
e) OR (compression stocking*) OR (compression sleeve*)
f ) AND (pneumatic compression) OR (compression pump)
AND
g) (reduc* limb volume)
h) OR (reduc* limb size)
i) OR (reduc* excess volume)
j) OR (reduc* excess limb volume)
k) OR (reduc* oedema OR edema volume)
Lymph?edema OR elephantiasis
l) AND physical therapy
m) AND bandage*
n) AND (compression bandage*)
o) AND hosiery OR hose
p) AND (compression hosiery OR hose)
q) AND (compression stocking*) OR (compression sleeve*)
AND
r) (Quality of Life measure*) OR (Quality of Life tool*)
34Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
W H A T ’ S N E W
Last assessed as up-to-date: 19 February 2008.
8 May 2008 New search has been performed Updated 2008 as part of the Cochrane Updating Project. New search no change
to conclusions or citation. Risk of bias tables added
8 May 2008 Amended Converted to new review format.
H I S T O R Y
Protocol first published: Issue 3, 2001
Review first published: Issue 4, 2004
30 August 2004 Amended First review publication
30 May 2001 Amended Protocol publication
C O N T R I B U T I O N S O F A U T H O R S
CB was the principal reviewer and wrote up the review. CB and NP were responsible for the extraction of the data. NP acted as the
second reviewer and collated the final draft of the review. PM provided clinical direction and both KS and PM advised, and helped
write both the protocol and the review.
In 2008, Anne Lethaby undertook to update the review as part of the Cochrane updating project. A new search was conducted and trials
identified were independently scanned by Anne Lethaby and Julie Brown. No new trials were identified, but a number of potentially
relevant trials were added to the Excluded Studies tables and two new ongoing trials were identified and documented. Anne Lethaby
also assessed the included studies for risk of bias.
D E C L A R A T I O N S O F I N T E R E S T
CB and PM are authors on one of the included trials. CB was the principal reviewer and wrote up the review.
S O U R C E S O F S U P P O R T
35Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Internal sources
• Royal College of Nursing, UK.
External sources
• No sources of support supplied
N O T E S
The review was updated in 2008 as part of a pilot Cochrane updating project. No further trials met the inclusion criteria. A number
of trials were added to the Excluded Studies Table and a Risk of Bias table was added.
I N D E X T E R M S
Medical Subject Headings (MeSH)
∗Physical Therapy Modalities; Bandages; Lymphedema [∗rehabilitation]; Randomized Controlled Trials as Topic
MeSH check words
Humans
36Physical therapies for reducing and controlling lymphoedema of the limbs (Review)
Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.